kernel-agent-overview

Guide kernel generation workflows from user intent to validated results.

258|48|Updated Jun 22, 2020
One-click install
npx skills add https://github.com/mindspore-ai/akg --skill kernel-agent-overview-mindspore-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kernel-agent-overview
Source: https://github.com/mindspore-ai/akg/tree/main/akg_agents/python/akg_agents/op/resources/skills/kernel-agent/kernel-agent-overview
Command: npx skills add https://github.com/mindspore-ai/akg --skill kernel-agent-overview-mindspore-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

## What problem does it solve? KernelAgent 作为一个端到端的工作流指南,定义了从需求理解到任务描述生成、到执行与结果评估的完整流程,并在各阶段提供明确的用户确认点以确保可控性和可追溯性。

## Core Features & Use Cases

  • 提供分析用户需求、识别输入、生成 task_desc、以及执行方式确认的端到端流程。
  • 支持在 kernel 生成、验证、优化等场景中的协作,包含错误处理与回滚策略,确保高可靠性。
  • 典型用例包括将一个算子从需求转化为可执行代码、进行验证并提供改进建议。

### Quick Start 请按照流程分析需求、生成任务描述并在执行前获得用户确认后继续。

Frequently Asked Questions about kernel-agent-overview

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate end-to-end kernel generation safely from user intent?

Kernel generation automation uses a structured workflow to guide projects from user intent to validated results. It enforces strict prompts, guardrails, execution orchestration, and clear reporting to ensure reproducible outcomes.

What is the best way to manage task_desc creation for kernel code generation?

Managing task_desc creation involves analyzing user requirements and identifying inputs to generate a structured task description. The workflow requires user confirmations before proceeding to execution, ensuring controllability and traceability.

Can I use AI-assisted prompt engineering for operator validation and optimization?

AI-assisted prompt engineering supports operator validation and optimization tasks. It provides a collaborative workflow with error handling and rollback strategies to ensure high reliability across design and validation scenarios.

Does the kernel generation workflow require user confirmation before executing tasks?

The kernel generation workflow requires explicit user confirmation before executing tasks. This structured approach includes input analysis and task_desc creation stages to ensure safe execution and project traceability.

How to handle errors and rollback strategies during kernel generation?

Handling errors during kernel generation involves structured error handling and rollback strategies within the workflow. This ensures safe execution across design, validation, and optimization tasks to maintain high reliability.

Why do I need user confirmations in an automated AI-assisted workflow?

User confirmations are needed in an automated AI-assisted workflow to enforce guardrails and maintain controllability. They ensure safe execution and traceability by validating task descriptions before proceeding with code generation.